This inventor holds 1 USPTO granted patent. Top assignees: Korea Advanced Institute of Science and Technology, Soongsil University. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
1 out of 2,792 
1 out of 320 
Years Active: 2026
Title: Taehyung Kwon: Inventor in Neural Tensor Compression.
Introduction
Taehyung Kwon is an inventor associated with Daejeon, South Korea. His patent work includes a method for lossy tensor compression using neural network-based tensor-train decomposition. The available information lists 1 patent for him.
Latest Patents
“Lossy Tensor Compression Method Using Neural Network-Based Tensor-Train Decomposition” describes a computer-performed approach to lossy tensor compression. The method inputs the mode indices of a target entry into a neural tensor-train decomposition (NTTD) model and uses the model to obtain tensor-train (TT) cores for reconstructing that entry.
Career Highlights
Kwon is associated with the Korea Advanced Institute of Science and Technology and Soongsil University. His patent reflects work at the intersection of neural networks, tensor decomposition, and data compression.
Collaborations
Kijung Shin and Jihoon Ko are identified as Kwon’s coworkers.
Conclusion
Taehyung Kwon’s listed invention applies neural tensor-train decomposition to lossy tensor compression, connecting his work with research in machine learning and data representation.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile